A good support prompt isn't a magic trick. It's a checklist you don't have to remember, run by something that never gets bored.
Reuse the recipe
Lock the parts that never change and leave one slot for the variable. Now every run shares a family resemblance instead of starting from scratch.
- Save the prompt that worked so the next person reuses it
- Compare two outputs instead of trusting one
- Name the audience and the single goal before you generate
Generate options, not answers
Treat the first output as a draft, not a verdict. The prompt does the typing; you keep the judgment about what ships.
Ask for the reasoning, then the result. When the model shows its steps you can see exactly where to push back.